This post outlines a problem: You're doing it wrong: A plea for sanity in the Low Quality Posts queue

However, posting to meta isn't going to solve the issue, in my opinion. The majority of reviewers don't read meta posts.

Nor will people read the instructions in the queue if they get updated.

The solution here is to update the UI itself.

The current UI is just four buttons. "Looks OK", "Edit", "Recommend Deletion", and "Skip".

The issue is that people don't know which post goes in which bucket, and just assume -- many a time assuming wrong. Given that opinions on what should and shouldn't be deleted have always been divided (example), it's reasonable to predict that many of the reviewers will have a different picture in mind.

I propose a different solution: Provide more options that focus on what the post is, not on what should be done with it. Internally, it's fine if the buttons actually do the same thing. Using the categories from Undo's post, we could have the following buttons:

Looks good Partial answer Not an answer Link only answer Needs improvement Wrong answer Skip

The first two would internally equal a "looks good" vote, the next two equal a "Recommend Deletion" (with slightly tweaked, more specific comments), "Needs Improvement" will fire up an edit box (which, if canceled without any edits, will focus on the comment box with a suggestion to comment), and "Wrong answer" will auto-downvote, and focus on the comment box with a suggestion to explain why it's wrong.

(I do see that these are a lot of buttons to have, perhaps we could reduce them a bit)

We have precedent for this kind of "illusion of choice" UI. It may not have been deliberate, but the "off topic" options in the flag-to-close interface all do the same thing; they raise a generic "off topic" flag (with the possible exception of the migration flag; I forget). Even from the moderator UI it isn't possible to distinguish which exact flag was raised, and I suspect they're all stored in the database as the same thing.

But the choice there provides a very useful function. Since we force users to choose, they will generally read the options and decide what type of post it is. If it doesn't belong to those types, they will be less ready to flag and might put some thought into it. When the close reasons were updated to have the multiple "off topic" choices which had clear descriptions, this was one of the advantages that I loved about the new system -- it forced people to think, instead of work off of the vague idea of "This post, me no likey".

Of course, we should supplement these options with an explanation of what they do in the tips box shown in the review queue, to avoid confusion. But by having people categorize the type of question, rather than what they think should be done with it, we might be able to solve the issue of the LQ queue.

This post outlines a problem: You're doing it wrong: A plea for sanity in the Low Quality Posts queue

However, posting to meta isn't going to solve the issue, in my opinion. The majority of reviewers don't read meta posts.

Nor will people read the instructions in the queue if they get updated.

The solution here is to update the UI itself.

The current UI is just four buttons. "Looks OK", "Edit", "Recommend Deletion", and "Skip".

The issue is that people don't know which post goes in which bucket, and just assume -- many a time assuming wrong. Given that opinions on what should and shouldn't be deleted have always been divided (example), it's reasonable to predict that many of the reviewers will have a different picture in mind.

I propose a different solution: Provide more options that focus on what the post is, not on what should be done with it. Internally, it's fine if the buttons actually do the same thing. Using the categories from Undo's post, we could have the following buttons:

Looks good Partial answer Not an answer Link only answer Needs improvement Wrong answer Skip

The first two would internally equal a "looks good" vote, the next two equal a "Recommend Deletion" (with slightly tweaked, more specific comments), "Needs Improvement" will fire up an edit box (which, if canceled without any edits, will focus on the comment box with a suggestion to comment), and "Wrong answer" will auto-downvote, and focus on the comment box with a suggestion to explain why it's wrong.

(I do see that these are a lot of buttons to have, perhaps we could reduce them a bit)

We have precedent for this kind of "illusion of choice" UI. It may not have been deliberate, but the "off topic" options in the flag-to-close interface all do the same thing; they raise a generic "off topic" flag (with the possible exception of the migration flag; I forget). Even from the moderator UI it isn't possible to distinguish which exact flag was raised, and I suspect they're all stored in the database as the same thing.

But the choice there provides a very useful function. Since we force users to choose, they will generally read the options and decide what type of post it is. If it doesn't belong to those types, they will be less ready to flag and might put some thought into it. When the close reasons were updated to have the multiple "off topic" choices which had clear descriptions, this was one of the advantages that I loved about the new system -- it forced people to think, instead of work off of the vague idea of "This post, me no likey".

Of course, we should supplement these options with an explanation of what they do in the tips box shown in the review queue, to avoid confusion. But by having people categorize the type of question, rather than what they think should be done with it, we might be able to solve the issue of the LQ queue.

This post outlines a problem: You're doing it wrong: A plea for sanity in the Low Quality Posts queue

However, posting to meta isn't going to solve the issue, in my opinion. The majority of reviewers don't read meta posts.

Nor will people read the instructions in the queue if they get updated.

The solution here is to update the UI itself.

The current UI is just four buttons. "Looks OK", "Edit", "Recommend Deletion", and "Skip".

The issue is that people don't know which post goes in which bucket, and just assume -- many a time assuming wrong. Given that opinions on what should and shouldn't be deleted have always been divided (example), it's reasonable to predict that many of the reviewers will have a different picture in mind.

I propose a different solution: Provide more options that focus on what the post is, not on what should be done with it. Internally, it's fine if the buttons actually do the same thing. Using the categories from Undo's post, we could have the following buttons:

Looks good Partial answer Not an answer Link only answer Needs improvement Wrong answer Skip

The first two would internally equal a "looks good" vote, the next two equal a "Recommend Deletion" (with slightly tweaked, more specific comments), "Needs Improvement" will fire up an edit box (which, if canceled without any edits, will focus on the comment box with a suggestion to comment), and "Wrong answer" will auto-downvote, and focus on the comment box with a suggestion to explain why it's wrong.

(I do see that these are a lot of buttons to have, perhaps we could reduce them a bit)

We have precedent for this kind of "illusion of choice" UI. It may not have been deliberate, but the "off topic" options in the flag-to-close interface all do the same thing; they raise a generic "off topic" flag (with the possible exception of the migration flag; I forget). Even from the moderator UI it isn't possible to distinguish which exact flag was raised, and I suspect they're all stored in the database as the same thing.

But the choice there provides a very useful function. Since we force users to choose, they will generally read the options and decide what type of post it is. If it doesn't belong to those types, they will be less ready to flag and might put some thought into it. When the close reasons were updated to have the multiple "off topic" choices which had clear descriptions, this was one of the advantages that I loved about the new system -- it forced people to think, instead of work off of the vague idea of "This post, me no likey".

Of course, we should supplement these options with an explanation of what they do in the tips box shown in the review queue, to avoid confusion. But by having people categorize the type of question, rather than what they think should be done with it, we might be able to solve the issue of the LQ queue.

This post outlines a problem: You're doing it wrong: A plea for sanity in the Low Quality Posts queue

However, posting to meta isn't going to solve the issue, in my opinion. The majority of reviewers don't read meta posts.

Nor will people read the instructions in the queue if they get updated.

The solution here is to update the UI itself.

The current UI is just four buttons. "Looks OK", "Edit", "Recommend Deletion", and "Skip".

The issue is that people don't know which post goes in which bucket, and just assume -- many a time assuming wrong. Given that opinions on what should and shouldn't be deleted have always been divided (example), it's reasonable to predict that many of the reviewers will have a different picture in mind.

I propose a different solution: Provide more options that focus on what the post is, not on what should be done with it. Internally, it's fine if the buttons actually do the same thing. Using the categories from Undo's post, we could have the following buttons:

Looks good Partial answer Not an answer Link only answer Needs improvement Wrong answer Skip

The first two would internally equal a "looks good" vote, the next two equal a "Recommend Deletion" (with slightly tweaked, more specific comments), "Needs Improvement" will fire up an edit box (which, if canceled without any edits, will focus on the comment box with a suggestion to comment), and "Wrong answer" will auto-downvote, and focus on the comment box with a suggestion to explain why it's wrong.

(I do see that these are a lot of buttons to have, perhaps we could reduce them a bit)

We have precedent for this kind of "illusion of choice" UI. It may not have been deliberate, but the "off topic" options in the flag-to-close interface all do the same thing; they raise a generic "off topic" flag (with the possible exception of the migration flag; I forget). Even from the moderator UI it isn't possible to distinguish which exact flag was raised, and I suspect they're all stored in the database as the same thing.

But the choice there provides a very useful function. Since we force users to choose, they will generally read the options and decide what type of post it is. If it doesn't belong to those types, they will be less ready to flag and might put some thought into it. When the close reasons were updated to have the multiple "off topic" choices which had clear descriptions, this was one of the advantages that I loved about the new system -- it forced people to think, instead of work off of the vague idea of "This post, me no likey".

Of course, we should supplement these options with an explanation of what they do in the tips box shown in the review queue, to avoid confusion. But by having people categorize the type of question, rather than what they think should be done with it, we might be able to solve the issue of the LQ queue.